J. Embryol. exp. Morph. 75, 225-239 (1983) 2 2 5Printed in Great Britain © The Company of Biologists Limited 1983

Localization of messenger RNA in the cortex of

Chaetopterus eggs and.early embryos

By WILLIAM R. JEFFERY1 AND LINDA J. WILSONFrom the Department of Zoology, University of Texas, Austin, and the Marine

Biological Laboratory, Woods Hole, Massachusetts

SUMMARY

The distribution of mRNA in Chaetopterus pergamentaceus eggs was examined by in situhybridization with poly(U) and specific cloned DNA probes. Eggs contain three distinctregions; the cortical ectoplasm, endoplasm, and a plasm released from the germinal vesicle(GV) during maturation. The ectoplasm of the mature egg showed a 15-fold enrichment inpoly(A) and in histone and actin mRNAs relative to the endoplasm and the GV plasm afterin situ hybridization. More than 90% of the total mass of egg poly (A) + RNA and histone andactin messages was estimated to be present in the ectoplasm. The mRNA molecules co-distributed with ectoplasmic inclusion granules during ooplasmic segregation. During theextensive cytoplasmic rearrangements which occur at the time of the first cleavage theectoplasm was divided into animal and vegetal portions. The animal portion was segregatedevenly between the AB and CD blastomeres, whereas the vegetal portion entered the polarlobe and was preferentially segregated to the CD blastomere. Histone and actin mRNAentered both the AB and CD blastomeres of the 2-cell embryo. The results demonstrate thatmRNA is quantitatively localized in the cortex of the Chaetopterus egg and early embryo.

INTRODUCTION

The localization of maternal mRNA molecules and their segregation to dif-ferent embryonic cells has been proposed to mediate cell determination duringearly development (see Davidson, 1976 and Jeffery, 1983 for reviews). Theexistence of localized maternal mRNA molecules, however, is still a controver-sial issue in most embryonic systems. Solution hybridization studies have un-covered differences in the complexity of poly(A)+ RNA molecules located indifferent regions of sea urchin (Rodgers & Gross, 1978; Ernst et al. 1980) andXenopus laevis (Carpenter & Klein, 1982) embryos. Qualitative differences,however, are yet to be detected between the in vitro translation products ofprevalent poly(A)+ RNA molecules isolated from different parts of embryos(Ilyanassa; Brandhorst & Newrock, 1981; Collier & McCarthy, 1981), suggest-ing that the localized RNA sequences detected by hybridization are either veryrare or do not serve as messages. Conflicting results were also obtained when the

1 Author's address: Department of Zoology, University of Texas, Austin, Texas 78712,U.S.A.

226 W. R. JEFFERY AND L. J. WILSON

spatial distribution of total poly(A)+ RNA was examined by in situ hybridiza-tion with poly(U). Poly(A)+ RNA sequences were found to be evenlydistributed in sea urchin (Angerer & Angerer, 1981) and mouse (Sternlicht &Schultz, 1981; Piko & Clegg, 1982) embryos. In contrast, poly(A)+ RNA ap-peared to be localized when other embryos, particularly mosaic embryos, wereanalysed by in situ hybridization with poly(U). For instance, about half the massof poly (A) + RNA was present in the plasm derived from the germinal vesicle(GV) of Styela eggs and remained localized in this region after maturation,ooplasmic segregation and cleavage (Jeffery & Capco, 1978). The localizedpoly(A)+ RNA of Styela embryos was primarily segregated to the animalhemisphere blastomeres during early embryogenesis. Localizations of poly (A)+RNA have also been reported in the cortex of Oncopeltus fasciatus (Capco &Jeffery, 1979) and Xenopus laevis (Capco & Jeffery, 1982) oocytes after in situhybridization.

In the present study we have continued the comparative analysis of the spatialdistribution of mRNA during early embryonic development. The egg ofChaetopterus has been selected for further study because, like that of Styela, itcontains a GV plasm and distinct cytoplasmic regions which are subject toooplasmic segregation and differential partitioning between the embryonic cellsduring early development (Lillie, 1906). In situ hybridization with poly(U) andcloned DNA probes for specific messages has revealed that most of the mRNAof eggs and early embryos is localized in the cortex.

MATERIALS AND METHODS

Chaetopterus pergamentaceus was obtained from the Marine ResourcesDepartment of the Marine Biological Laboratory, Woods Hole, MA. Adults,gametes and embryos were maintained and handled by established procedures(Costello & Hendley, 1971). Parapodia, oocytes and embryos at the desired stageof development were fixed at -20°C for 30min in Petrunkewitsch's fluid. Thisfixative, which we have found to quantitatively preserve poly(A)+ RNA in thespecimens, was freshly prepared before use by mixing one part of an aqueoussolution containing 12 % nitric acid (v/v)-8 % cupric nitrate (w/v) with threeparts of an aqueous solution containing 76 % ethanol (v/v), 5-7 % ethyl ether(v/v) and 3-8 % crystalline phenol (w/v). The fixed specimens were dehydratedin ethanol and cleared in toluene at -20 °C. They were embedded in paraplast,sectioned at 8 fim and attached to glass slides for histology or in situ hybridization.

In situ hybridization with [3H]poly(U) [4-65 Ci/mmole; New EnglandNuclear, Boston, MA.] was carried out as described previously (Capco & Jef-fery, 1978), except that a step involving treatment of the slides with 10/ig/mlproteinase K (lOmin at 20 °C) was inserted between the DNase I pretreatmentand the annealing. This step was necessary to obtain the highest efficiency ofpoly(A) + RNA detection in sections of Chaetopterus eggs.

Cortical mRNA 227The DNA probes were prepared by Hind III restriction of plasmids containing

the Dm-500 histone gene complex (Lifton et al. 1977) or the Dm-A2 actin gene(Fyrberg etal. 1980) from Drosophila melanogaster. The 4-6 kilobase (kb) DNAfragment of the Dm-500 complex contains complete sequences of the genes forHI, H2a, H2b, H3, and H4 and their spacers (Lifton et al. 1977) and was usedas a probe for histone mRNA. The 1-8 kb DNA fragment of the Dm-A2 actingene contains almost the entire coding sequence of an actin mRNA (Fyrberg etal. 1980) and was used as a probe for actin mRNA. The Hind III digests wereseparated by agarose gel electrophoresis and bands containing the appropriateDNA fragments were excised from the gel. The DNA was extracted from the gelslices and nick translated with [125I]deoxycytidine triphosphate (2-5 x l t^Ci/mmole; New England Nuclear, Boston, MA) to a specific activity of about1-5 x 107d.p.m./^g DNA as described by Maniatis et al. (1975). The DNAprobes were dissolved in hybridization buffer, denatured by heating at 90 °C for5 min and applied to the sections at saturating concentrations (2-10ng/ml; 20 [Aper slide). In situ hybridization with cloned histone and actin DNA probes wascarried out according to the method of Jeffery (1982).

The autoradiographs and sections for cytological observation were stainedwith Harris haematoxylin-eosin for 2-5 min. This stains the ectoplasmic granulesred, the endoplasm and yolk granules light blue, and the GV plasm dark purple.

RESULTS AND DISCUSSION

Behaviour of cytoplasmic regions during early development

The cytoplasmic regions of Chaetopterus eggs and their movements duringmaturation and early embryogenesis were originally described by Lillie (1906).Lillie's published account did hot consider the behaviour of the cytoplasmicregions during the entire period between maturation and the first cleavage. Thusit was necessary to extend Lillie's cytological description of early Chaetopterusdevelopment before examining the distribution of mRNA during early develop-ment. A summary of Lillie's and our own cytological observations is presentedin Figs 1-8.

Chaetopterus eggs are released from the parapodia as primary oocytes contain-ing three regions, the GV plasm, endoplasm and ectoplasm (Fig. 1). Theendoplasm consists of large blue-staining, yolk particles and clusters of smaller,red-staining granules. The ectoplasm, situated in the animal two-thirds of theoocyte cortex, is densely packed with these same red-staining granules (ectoplas-mic granules). Within 10 min after the oocytes are exposed to sea water the GVruptures initiating maturation and ooplasmic segregation (Fig. 2). The GV plasmmoves into the animal pole region where an ectoplasmic defect (Lillie, 1906)appears at the place where the polar bodies are formed. Simultaneously, theclusters of internal ectoplasmic granules move from the endoplasm into thecortex and join with the rest of the ectoplasm to form a cortical layer which

228 W. R. JEFFERY AND L. J. WILSON

ec

Figs 1-8

Cortical mRNA 229surrounds the entire surface of the egg, except for the region of the ectoplasmicdefect (Figs 2, 3).

Cytoplasmic rearrangements also occur in fertilized eggs during cleavage andpolar lobe formation. The cleavage associated movements are first detected atmetaphase when the egg elongates into a pear-shaped mass with the narrowestportion of the cell in the animal hemisphere (Fig. 4). Many of the ectoplasmicgranules are temporarily dislodged from the cortex at this time and accumulateat the boundary between the GV plasm and the endoplasm (Fig. 4). The pear-shaped cell is generated by a cortical constriction which begins to form in theanimal hemisphere and proceeds vegetally, eventually resulting in polar lobeprotrusion (Figs 5,6). Prior to polar lobe formation the ectoplasmic granulesreturn to the cortex and are split into animal and vegetal fields by the advancingcortical constriction (Fig. 5). At cleavage the animal ectoplasmic field is dividedabout equally between the AB and CD blastomeres (Figs 6-8). In contrast, mostof the vegetal ectoplasmic field, along with the other contents of the polar lobe,enter the CD blastomere (Figs 6, 7). After polar lobe regression at telophase itsectoplasmic granules accumulate in the vegetal region of the constrictingcleavage furrow and become localized in the cortex of the CD blastomere (Figs7,8). Although we have not examined the fate of the vegetal ectoplasmic fieldbeyond the 2-cell stage, according to Lillie (1906) it is also present in the secondpolar lobe and is thus likely to enter the D quadrant of the embryo.

Localization of mRNA in the cortical ectoplasm

The distribution of grains in mature eggs after in situ hybridization withpoly(U) is shown in Fig. 9 and quantified in Table 1. The grains were con-centrated about 15-fold in the cortical ectoplasm relative to the GV plasm or theendoplasm. The interaction of poly(U) with the eggs was substantially reduced

Figs 1-8. The behaviour of cytoplasmic regions during maturation and earlyembryogenesis of Chaetopterus pergamentaceus eggs. These drawings represent sec-tions through the animal-vegetal axis of fixed and stained oocytes, eggs and embryos.They are drawn in the style of Lillie (1906) and are a composite of Lillie's and ourown histological studies. Fig. 1. A primary oocyte as it appears shortly after itsrelease into sea water. Fig. 2. An egg at the first maturation division. Fig. 3. An eggwhich has completed the first maturation division. Fig. 4. A zygote at the pear stageshowing the transient accumulation of cortical ectoplasmic granules at the boundarybetween the endoplasm and the residual germinal vesicle plasm. Fig. 5. A zygote atmetaphase showing the equatorial constriction which splits the ectoplasm into animaland vegetal portions. Fig. 6. A trefoil embryo showing the accumulation of vegetalectoplasm in the polar lobe. Fig. 7. A telophase embryo showing the accumulationof vegetal ectoplasmic granules along one side of the vegetal cleavage furrow follow-ing the retraction of the polar lobe. Fig. 8. A two-cell embryo showing thedistribution of animal and vegetal ectoplasms in the AB (right) and CD (left) blasto-meres. Germinal vesicle or residual plasm of the germinal vesicle (GV); endoplasm(en), ectoplasm (ec); polar lobe (PL). The filled spheres represent the ectoplasmicgranules. The open spheres represent the endoplasmic yolk platelets.

230 W. R. JEFFERY AND L. J. WILSON

era

Figs 9, 10. /n situ hybridization of Chaetopterus eggs with poly(U). Fig. 9. Anautoradiograph of a mature egg showing grains in the cortical ectoplasm. X400. Fig.10. An autoradiograph of a mature egg treated with RNase T2 prior to in situhybridization. Cortical grains are not apparent. x400. See Table 1 for details of theRNase treatment. Ectoplasm (ec).

when the sections were treated with RNase (Fig. 10; Table 1) or alkali (data notshown) prior to in situ hybridization suggesting that the probe binds to RNA inthe section. It is unlikely that the concentration of mRNA in the cortex is dueto the differential extraction of RNA during fixation since we have shown thatpoly(A)+ RNA is quantitatively retained in eggs fixed with Petrunkewitsch'sfluid. When sections of eggs were subjected to in situ hybridization with thehistone or actin DNA probes, the grain concentration also ranged between 10-and 15-fold higher in the ectoplasm than in the other cytoplasmic regions (Table1; Figs 19 and 21). To estimate the proportion of the egg poly(A) present in thecortical ectoplasm the grain counts were multiplied by the volume comprised byeach of the cytoplasmic regions (as derived from their areas in egg sections;Jeffery & Capco, 1978) and these values were expressed as percentages of thetotal grains. Although the ectoplasm comprises only 35 % of the egg volume, itwas estimated to contain about 95 % of the mass of egg poly(A). These resultsindicate that a population of mRNA molecules, comprising most of the eggpoly(A)+ RNA, histone mRNA and actin mRNA, is localized in the corticalectoplasm.

The case of Chaetopterus now represents the fourth example of quantitativemRNA localization during early development. In vitellogenic oocytes of Onco-peltus poly(A)+ RNA is present in the anterior and posterior cortical cytoplasms

Cortical mRNA 231

Table 1. In situ hybridization of Chaetopterus eggs with [3H]poly(U) andI125-labelled actin and histone DNA probes

Probe and Pretreatment

Poly(U)Poly(U), RNase T2Histone DNAHistone DNA, RNase AActin 1-8 kb. DNAActin 1-8 kb. DNA, RNase ApBR322

Ectoplasm

62-3 ±2-13-5 ±1-3

72-6 ±9-23-7 ±0-8

42-4 ±3-33-9 ±1-62-4 ±1-0

Grain Number ± S.D.

Endoplasm

3-7 ±0-93-1 ±1-48-112-51-1 ±0-64-611-30-42-2 ±0-7

ResidualGV plasm

0-51-2 ±0-31-7 ±0-90-52-5 ±1-30-22-5 ±1-6

Grain numbers are expressed as the mean of counts in 15-25 different 100 ̂ m2 areas ±standard deviation (S.D.). The indicated slides were pretreated with 50jUg/ml pancreaticRNase A for 24 h at 37 °C or 50^g/ml RNase T2 for 24 h at 37 °C. The RNases were dissolvedin lOmM-Tris-HCl (pH7-6), 50mM-KCl, lmM-MgCl2.

(Capco & Jeffery, 1979). In stage-6 oocytes of Xenopus laevis poly(A)+ RNAis localized in the cortex of the animal hemisphere (Capco & Jeffery, 1982). Inthe latter two cases localization of poly(A)+ RNA is a transient phenomenonand cannot be detected in the mature egg. In Styela about half of the poly(A)+RNA mass is concentrated in the GV plasm of oocytes and after maturation itbecomes localized in the ectoplasm (Jeffery & Capco, 1978). The lack of asignificant accumulation of poly(A) + RNA in the GV plasm of Chaetopterus orXenopus (Capco & Jeffery, 1982) oocytes suggests that this feature is not ageneral developmental phenomenon. It is also notable that in three of the fourcases examined mRNA localizations are present in the egg cortex, a finding ofinterest in light of the morphogenetic significance ascribed to this area of the egg.

Origin of cortical mRNA

Two major possibilities exist for the origin of the cortical poly(A)+molecules. They could be present in the cortex of the primary oocyte beforematuration or they could migrate into the egg cortex from sites such as the GVplasm or the endoplasm during ooplasmic segregation. To resolve this issueprimary oocytes were fixed and subjected to in situ hybridization with poly(U)during the period between their release from the parapodia and the completionof maturation. As shown in Fig. 11 grains were already concentrated in thecortex of the primary oocyte. The co-distribution of these grains with thecortical ectoplasmic granules is demonstrated by sections through the animal-vegetal axis showing heavy labelling in the animal two-thirds of the cortex (Fig.13). Very few grains were observed in the GV plasm of the primary oocyte orthe mature egg (Figs 11-14). In contrast to the mature egg, however, the

232 W. R. JEFFERY AND L. J. WILSON

primary oocyte also showed significant labelling over specific areas of theendoplasm (compare Figs 11 and 12 to Fig. 13). The endoplasmic grains werealways positioned immediately above clusters of ectoplasmic granules (Fig. 1).Two factors appear to contribute to the localization of mRNA in the egg cortex.The bulk of the oocyte mRNA must already be present in the cortical ectoplasmprior to maturation while the remainder is likely to be incorporated into thecortex with the internal ectoplasmic granules during ooplasmic segregation. Thecortical mRNA localization probably originates early during oogenesis since itcan be detected in parapodial oocytes of all sizes (Fig. 14).

Segregation of cortical mRNA during cleavage

Sections of fertilized eggs were fixed at intervals during the first cleavage andsubjected to in situ hybridization with poly(U) to determine the fate of thecortical poly (A) + RNA during the unequal partitioning of ectoplasm betweenthe AB and CD blastomeres. The results are shown in Figs 15-18. Pear-stageembryos showed grains concentrated in the cortical ectoplasm and in the vicinityof ectoplasmic granules that were dislodged from the cortex and accumulated atthe edge of the residual GV plasm (Fig. 15). A few grains were still present atthis time over the ectoplasmic spherules that remain at the egg surface. Later,when the polar lobe begins to form, heavy labelling was seen over accumulationsof ectoplasmic granules in the vegetal pole region of the egg. Grains were con-centrated in two major locations at the trefoil stage, the cortical ectoplasm of theanimal hemisphere and the polar lobe (Fig. 16). After the polar lobe is retractedinto the nascent CD cell at telophase and the vegetal ectoplasmic granules moveinto the region of the cleavage furrow (Figs 7 and 8), intense labelling was seenin the furrow region directly above these granules (Fig. 17). The population ofmRNA molecules in the animal ectoplasmic field appears to be divided aboutequally between the AB and CD blastomeres. In contrast, most of the mRNAmolecules in the vegetal ectoplasmic field enter the polar lobe and are preferenti-ally distributed to the CD blastomere (Fig. 18). The results indicate that thecortical mRNA molecules are co-distributed with ectoplasmic granules duringthe first cleavage as well as ooplasmic segregation.

Two previous studies suggest that the position of mRNA molecules in the egg

Figs 11-14. In situ hybridization of Chaetopterus oocytes with poly(U). Fig. 11. Anautoradiograph of a recently shed primary oocyte showing abundant grains over theectoplasm (ec) and few grains over the germinal vesicle (GV). Clusters of grains alsoappear in the endoplasm (en). X400. Fig. 12. An autoradiograph of a primary oocyteshowing clusters of grains (arrows) over ectoplasmic granules in the endoplasm.xlOOO. Fig. 13. An autoradiograph of a primary oocyte sectioned through theanimal-vegetal axis showing abundant grains in the animal two-thirds of theectoplasm and few grains in the endoplasm and residual GV plasm. Animal pole,(AP). x400. Fig. 14. An autoradiograph of parapodial oocytes of various sizesshowing cortical ectoplasmic grains. X150.

Cortical mRNA 233

G.¥

\rrt

234 W. R. JEFFERY AND L. J. WILSON

A^£*I*

«•

Figs 15, 16. /n situ hybridization of Chaetopterus embryos with poly(U) during theperiod immediately prior to the first cleavage. Fig. 15. An autoradiograph of anequatorial section through a pear stage embryo showing grains over ectoplasmicgranules in the cortex and at the interface between the endoplasm (en) and theresidual GV plasm (GVP). x600. Fig. 16. An autoradiograph of a trefoil embryosectioned along the animal-vegetal axis with focus over the polar lobe. Grains areconcentrated mainly in the animal ectoplasm and in the cortex of the polar lobe. ABblastomere, (AB). CD blastomere, (CD). x400.

cytoplasm is fixed by associations with regionalized structures. First, the con-centration of mRNA does not appear to change in cytoplasmic regions whichmigrate extensively through the Styela egg during ooplasmic segregation (Jeffery

Cortical mRNA 235

18Figs 17, 18. Fig. 17. In situ hybridization of cleaving Chaetopterus embryos withpoly(U). An autoradiograph of a section through the animal-vegetal axis of a cleav-ing embryo showing the accumulation of grains over ectoplasmic granules in thevegetal region of the cleavage furrow. xlOOO. Fig. 18. An autoradiograph of a two-cell embryo showing the distribution of grains between the animal (AE) and vegetal(VE) ectoplasmic fields in the AB and CD blastomeres. x600.

236 W. R. JEFFERY AND L. J. WILSON

& Capco, 1978). Second, mRNA isolated from the vegetal pole region ofXenopus laevis eggs tends to accumulate in a vegetal pole to animal polegradient after microinjection into zygotes between fertilization and the firstcleavage (Capco & Jeffery, 1981). The localization of mRNA molecules in theectoplasm of Chaetopterus eggs could be due to an interaction with the cyto-skeletal elements known to reside in the cortex of many eggs (Franke et al.1976; Kidd & Mazia, 1980; Lehtonen & Badley, 1980; Colombo et al. 1981;Jeffery & Meier, 1983) or organelles associated with the cortical cytoskeleton.A possible candidate for the latter are the granular, nuage-like bodies recentlydescribed in the ectoplasm of Chaetopterus eggs (Eckberg, 1981).

Segregation of histone and actin mRNA sequences during cleavageThe partitioning of the cortical ectoplasm into animal and vegetal fields and

their unequal division between the AB and CD blastomeres during cleavagebrings up the possibility that specific mRNA sequences may be segregated intodifferent parts of the embryo. As an initial test of specific mRNA segregation insitu hybridization with histone and actin DNA probes was carried out on sectionsof eggs and cleaving embryos. The signal obtained was sensitive to pretreatmentof the sections with RNase and in situ hybridization with pBR322 failed togenerate a significant autoradiographic signal suggesting that DNA does notadventitiously bind to the sections (Table 1). Grains were found to be localizedin the cortical ectoplasm with little significant activity in the other regions of themature egg (Figs 19-22). In trefoil embryos the animal and the vegetal ectoplas-mic fields were labelled to the same extent (Figs 20, 22). Thus the localizationand segregation of cortical histone and actin messages is identical to that of thetotal poly(A) + RNA.

Several functional roles can be envisioned for mRNA localization during earlyembryonic development. First, the differential segregation of mRNA couldprovide metabolically-active cell lineages with an excess of maternal mRNA.Second, differential mRNA distribution might reflect a translational-levelcontrol mechanism in which messages are physically separated from the proteinsynthetic machinery (Showman et al. 1982). Third, some localized mRNAmolecules may be cytoplasmic morphogens which dictate the developmentalchoices made by embryonic cell lineages. At present we are unable to decidebetween these possibilities; it is possible all three roles may be played duringearly development. The Chaetopterus egg, however, provides an excellent sys-tem to test for the qualitative segregation of specific mRNA species because thecortical mRNA mass is split into two parts during cleavage and one part is largelydelivered to the CD blastomere. Although we have shown that the histone andactin messages are distributed to both the AB and CD cells, this result does notexclude the possibility that other mRNA species are differentially segregated.The distribution of many individual messages will have to be determined toassess the possibility of selective mRNA segregation.

Cortical mRNA 237

CD AB

7 -

AE

Figs 19-22. In situ hybridization of mature eggs and trefoil embryos with histone andactin DNA probes. Fig. 19. An autoradiograph of a mature egg treated with thehistone DNA probe showing grains concentrated in the cortical ectoplasm. X400.Fig. 20. An autoradiograph of a section through a trefoil stage embryo treated withthe histone DNA probe snowing grains in the animal hemisphere and the polar lobe.Fig. 21. An autoradiograph of a section through a mature egg treated with the actinDNA probe showing grains concentrated in the cortical ectoplasm. x400. Fig. 22.An autoradiograph of a section through a trefoil stage embryo treated with the actinDNA probe showing grains concentrated in the animal hemisphere and the polarlobe. x400. Animal ectoplasm, (AE); vegetal ectoplasm, (VE); AB blastomere,(AB); CD blastomere, (CD).

238 W. R. JEFFERY AND L. J. WILSON

Technical assistance was provided by Ms Dianne McCoig. The drawings were executed byMs Bonnie Brodeur. This research was supported by grant HD-13970 from the NationalInstitutes of Health.

REFERENCESANGERER,L. M. & ANGERER,R. C. (1981). Detection of poly(A)+RNA in sea urchin eggs and

embryos by quantitative in situ hybridization. Nucleic Acids Res. 9, 2819-2840.BRANDHORST, B. P. & NEWROCK, K. M. (1981). Post-transcriptional regulation of protein

synthesis in Ilyanassa embryos and isolated polar lobes. Devi Biol. 83, 250-254.CAPCO, D. G. & JEFFERY, W. R. (1978). Differential distribution of poly(A)-containing RNA

in the embryonic cells of Oncopeltus fasciatus: Analysis by in situ hybridization with a pH]-poly(U) probe. Devi Biol. 67, 137-151.

CAPCO, D. G. & JEFFERY, W. R. (1979). Origin and spatial distribution of maternal messengerRNA during oogenesis of an insect, Oncopeltus fasciatus. J. Cell Sci. 39, 63-76.

CAPCO, D. G. & JEFFERY, W. R. (1981). Regional accumulation of vegetal pole poly(A)+RNAinjected into fertilized Xenopus eggs. Nature 294, 255-257.

CAPCO, D. G. & JEFFERY, W. R. (1982). Transient localizations of messenger RNA in Xenopuslaevis oocytes. Devi Biol. 89, 1-12.

CARPENTER, C. D. & KLEIN, W. H. (1982). A gradient of poly(A)+RNA sequences in Xenopuslaevis eggs and embryos. Devi Biol. 91, 43-49.

COLLIER, J. R. & MCCARTHY, M. E. (1981). Regulation of polypeptide synthesis during earlyembryogenesis of Ilyanassa obsoleta. Differentiation 19, 31-46.

COLOMBO, R., BENEDUSI, P. & VALLE, G. (1981). Actin in Xenopus development: Indirectimmunofluorescence study of actin localization. Differentiation 20, 45-51.

COSTELLO, D. P. & HENLEY, C. (1971). Methods For Obtaining and Handling Marine Eggs andEmbryos. Woods Hole, MA.: Marine Biological Laboratory. p63-66.

DAVIDSON, E. H. (1976). Gene Activity in Early Development. New York: Academic Press.ECKBERG, W. R. (1981). An ultrastructural analysis of cytoplasmic localization in Chaetop-

terus pergamentaceus. Biol. Bull. 160, 228-239.ERNST, S., HOUGH-EVANS, B. R., BRITTEN, R. J. & DAVIDSON, E. H. (1980). Limited complex-

ity of the RNA in micromeres of sixteen-cell sea urchin embryos. Devi Biol. 79, 119-127.FRANKE, W. W., RATHKE, P. C , SIEB, E., TRENDELENBURG, M. F., OSBORN, M. & WEBER, K.

(1976). Distribution and mode of arrangement of microfilamentous structure and actin inthe cortex of the amphibian oocyte. Cytobiol. 14, 111-130.

FYRBERG, E. A., KINDLE, K. L., DAVIDSON, N. & SODJA, A. (1980). The actin genes ofDrosophila: A dispersed multigene family. Cell 19, 365-378.

JEFFERY, W. R. (1982). Messenger RNA in the cytoskeletal framework: Analysis by in situhybridization. J. Cell Biol. 95, 1-7.

JEFFERY, W. R. (1983). Maternal RNA and the embryonic localization problem. In: Controlof Embryonic Gene Expression, (ed. M. A. Q. Siddiqui). Boca Raton, FL: CRC Press.

JEFFERY, W. R. & CAPCO, D. G. (1978). Differential accumulation and localization of maternalpoly(A)-containing RNA during early development of the ascidian, Styela. Devi Biol. 67,152-166.

JEFFERY, W. R. & MEIER, S. (1983). A yellow crescent cytoskeletal domain in ascidian eggsand its role in early development. Devi Biol. 96, 125-143.

KIDD, P. & MAZIA, D. (1980). The ultrastructure of surface layers isolated from fertilized andchemically stimulated sea urchin eggs. /. Ultrastruct. Res. 70, 58-69.

LEHTONEN, E. & BADLEY, R. A. (1980). Localization of cytoskeletal proteins in pre-implantation mouse embryos. /. Embryol. exp. Morph. 55, 211-225.

LIFTON, R. P., GOLDBERG, M. L., KARP, R. W. & HOGNESS, D. S. (1977). The organizationof the histone genes in Drosophila melanogaster. Functional and evolutionary implications.ColdSpr. Harb. Symp. Quant. Biol. 42,1047-1051.

LILLIE, F. R. (1906). Observations and experiments concerning the elementary phenomenaof embryonic development in Chaetopterus. J. exp. Zool. 3,153-268.

Cortical mRNA 239MANIATIS, T., JEFFREY, A. & KLEID, D. (1975). Nucleotide sequence of the rightward operator

of phageA. Proc. natn Acad. ScL, U.S.A. 72, 1184-1188.PIKO, L. & CLEGG, K. B. (1982). Quantitative changes in total RNA, total poly(A), and

ribosomes in early mouse embryos. Devi Biol. 89, 362-378.RODGERS, R. E. & GROSS, P. R. (1978). Inhomogeneous distribution of egg RNA sequences

in the early embryo. Cell 14, 279-288.SHOWMAN, R. M., WELLS, D. E., ANSTROM, J., HURSH, D. A. & RAFF, R. A. (1982). Message-

specific sequestration of maternal histone mRNA in the sea urchin egg. Proc. natn Acad.Sci. U.S.A. 79, 5944-5947.

STERNLICHT, A. L. & SCHULTZ, R. M. (1981). Biochemical studies of mammalian oogenesis:Kinetics of accumulation of total and poly(A)-containing RNA during growth of mouseoocyte. /. exp. Zool. 215, 191-200.

(Accepted 16 February 1983)